Of substitutes and complements: trade credit versus bank loans in Japan, 1980-2012

This paper explores the interplay of trade credit and short-term bank loans between 1980 and 2012 in Japan. Many of the issues discussed for the U.S. remain largely unexplored for Japan, yet Japan is an interesting country study because of the particular features of its financial sector and its experience of extended zero interest rates. This paper makes three contributions: (1) We develop a new approach to using the Bank of Japan tankan survey, matched with individual firm-level data, to explore the impact of perceived business conditions and credit constraints on trade credit. (2) We divide the 30 years into sub-periods to gain a more granular view of the trade credit-bank loan relationship over time. (3) We use firm-level data to explore the relationship between trade credit and bank loans for listed firms in distress. We find that trade credit was a bank loan substitute during the upswing of the 1980s, but a complement (driven by the transaction motive) during the stagnation of the 1990s and 2000s. At the firm level, trade credit is contingent on the financial health of the borrower. For firms in distress, in the 1980s and 1990s trading partners extended trade credit after the main bank had come to the rescue. In contrast, in the early 2000s, both main banks and trading partners have reduced support of firms in distress, removing any substitution effect. Trade credit discriminates at the micro-level, which explains why previous research has yielded ambiguous results on the substitute versus complement question.